I am trying to extract group differences using dual regression ICA. I could find only a few articles about this and I could not thoroughly understand the way it works.
Could someone kindly explain how it works in simple form? thank you in advance.
Hi I am not expert in this topic but may be this is helpful to you.
Many large-scale functional brain networks have been shown to be reproducible across subjects . Dual Regression allows researchers to investigate individual and group differences in the structure of these networks, as well as to identify subject-specific networks based on networks identified at the group level.
Computation and Analysis Considerations
The following steps are performed for each spatial map provided for Dual Regression:
Spatial Regression: The spatial map is used as a spatial regressor to find the time series associated with the voxels in that map.
Temporal Regression: The time series found by spatial regression is then used as a temporal regressor to find the full set of voxels associated with that time series.
The result of these steps is a subject-specific spatial map based on the original spatial map.
The spatial maps used during Dual Regression can be generated any number of ways, but a common practice is to use Independent Component Analysis or clustering algorithms to generate group-specific maps of large-scale brain networks.The references you can go through.
Beckman et al, 2009
Filippini et al, 2009
Thank you very much every one. I do really appreciate your help. I am studying the articles and pages you recommended. and I find them really helpful. thank you.
Best Regards
Shiva
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3117074/#FD23: this is by far the best ref on the subject, as it explain the math details of the method and compares it with sevaral alternatives (GIGA3, etc.)
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression